A method for dynamic cardiac ultrasound simulation based on statistical ray tracing

Abstract

The present invention discloses a cardiac dynamic ultrasound simulation method based on statistical ray tracing, which belongs to the cross-technical field of computer graphics and medical simulation, and comprises S1: initializing world parameters of a physical simulation computing engine, and defining human tissue material information according to the physical properties of the heart, liver, kidney, blood vessel, lung, etc.; S2: defining scene file parameters and initializing the scene; S3: calculating the relative position and rotation of each joint in the cardiac model through the transformation matrix, obtaining each frame of grid information through the transformation matrix and importing it into the physical simulation computing engine, and updating the grid according to the time cycle; S4: the ultrasonic probe emits rays according to the sampling angle to perform statistical ray tracing; S5: generating an ultrasonic image through echoes; the present invention takes into account the scattering, absorption and propagation of sound waves in tissues, helps to simulate the formation of ultrasonic images, improves the physical accuracy of the simulation results, and realizes real-time simulation imaging of tissue and organ movement.

Description

Technical Field

[0001] The invention belongs to the cross-technical field of computer graphics and medical simulation, and specifically is a cardiac dynamic ultrasound simulation method based on statistical ray tracing. Background Art

[0002] Ultrasound imaging is a non-invasive imaging technology commonly used in medical diagnosis and biomedical research, which uses the propagation and reflection of high-frequency sound waves in tissues to generate images. This technology plays an important role in clinical diagnosis, surgical navigation and medical education.

[0003] Although traditional ray tracing algorithms can achieve real-time effects, the surfaces of tissues and organs are mostly uneven, and traditional ray tracing methods assume that the surfaces of tissues and organs are perfectly smooth, so the generated ultrasound images will cause sharp hard shadows. Statistical ray tracing is a powerful computer graphics technology that has been widely used to simulate the propagation and interaction of light in virtual scenes. Applying statistical ray tracing methods to ultrasound simulation can soften shadows, thereby simulating the ultrasound image generation process more realistically.

[0004] Previous ultrasound imaging methods only mentioned implementation solutions for static scenes, and did not optimize ultrasound simulation for complex dynamic organs such as the heart. Therefore, to solve the above problems, the present invention proposes a cardiac dynamic ultrasound simulation method based on statistical ray tracing to achieve dynamic ultrasound simulation of heart beating. Summary of the invention

[0005] The purpose of the present invention is to overcome the deficiencies of the prior art and provide a statistical ray tracing method for realizing dynamic ultrasound simulation.

[0006] The present invention proposes a cardiac dynamic ultrasound simulation method based on statistical ray tracing, and the steps are as follows:

[0007] Step S1: Initialize the bullet engine world parameters, and define the human tissue material information according to the physical properties of the heart, liver, kidney, blood vessels, lungs, muscles, bones, blood, and skin, wherein the human tissue material information includes acoustic impedance, attenuation coefficient, specular incidence, roughness, thickness, and expected and standard deviation of scattering;

[0008] Step S2: defining scene file parameters and initializing the scene according to the scene file, wherein the scene file parameters include the position of the ultrasound probe, resolution, pulse stretching rate, sampling angle, sampling number, pitch angle, frequency, transducer spacing, initial intensity of ultrasound, stop threshold, initial medium material of the light, imported model file and model scaling factor;

[0009] Step S3: a method for importing a heart animation model based on Bullet is proposed, wherein the relative position and rotation of each joint in the heart model are calculated through the transformation matrix, and the mesh information of each frame is obtained through the transformation matrix and imported into the Bullet engine, and the mesh is updated according to the time loop;

[0010] Step S31: creating a heart beating animation model;

[0011] Step S32: Use FBXSDK to parse the heart model animation and calculate and save the mesh of each frame:

[0012] ;

[0013] in, Indicates The frame index is The vertex coordinates of Indicates The frame index is The vertex global transformation matrix, Indicates that the index is The initial coordinates of the vertices;

[0014] Step S33: converting the animation frame grid into grid information that can be used by bullet;

[0015] ;

[0016] in, Indicates The frame index is The bullet world vertex position;

[0017] Step S34: importing mesh information into the physical world, and updating the mesh according to the specified time step;

[0018] Step S4: The ultrasonic probe emits rays according to the sampling angle to perform statistical ray tracing; generates random reflection and refraction directions according to the roughness of the surface material of the sound wave hitting point; modifies the hitting point position according to the thickness of the surface material of the hitting point, and generates ultrasonic echo, radio frequency echo, reflection and refraction, and speckle at the new hitting point, and updates the ultrasonic beam intensity at the same time;

[0019] Step S41: The radio frequency echo is composed of recorded ultrasonic echoes:

[0020] ;

[0021] Where t represents time, T represents the time of impact, o represents the source point, and P T represents the impact point, Ω represents the hemispherical surface, I t＞T(t,o) represents the echo contribution received by the source point starting from the intersection point at time T with a direction of ω;

[0022] In practice, it is calculated in a discrete way:

[0023] ;

[0024] Among them, I(r(P T ,ω i )) indicates that by T Towards ω i Directional ray contribution;

[0025] Step S42: The reflection and refraction represent the physical effect that occurs when the impact point reaches the boundary of two media. The normal direction is randomly generated according to the roughness to determine the reflection and refraction direction. At the same time, the reflection and refraction intensity is calculated based on Snell's law:

[0026] ;

[0027] ;

[0028] ;

[0029] ;

[0030] Where θ1 represents the reflection angle, θ2 represents the refraction angle, V i represents the normalized incident direction, N represents the normal direction, Z1 and Z2 represent the acoustic impedance of the two media respectively. V t , V r Indicates the direction of refraction and reflection;

[0031] The intensity update process of the sound wave is expressed as:

[0032] ;

[0033] ;

[0034] in, I i Indicates the intensity of the sound wave when it reaches the impact point. I r represents the intensity of the reflected wave, I t It represents the intensity of the refracted wave;

[0035] Step S43: The speckle is generated because countless sub-wavelength particles act as point wave sources to scatter in all directions; the scattering effect is calculated by the ultrasonic speckle convolution model:

[0036] ;

[0037] Among them, x represents the horizontal direction, y represents the axial direction, and z represents the elevation angle. represents the added random noise, r(x,y,z) represents the scattering intensity, g(x,y,z) represents the parameter function of the human tissue scatterer, and h(x,y,z) represents the frequency of the sampler (ultrasound probe). f c Gaussian envelope of the modulation;

[0038] ;

[0039] Among them, σ x , σ y , σ z represent the pulse stretching ratio in transverse, axial and elevation angles respectively;

[0040] Step S44: The ultrasonic beam intensity is updated. In addition to the reflection and refraction update, the ultrasonic intensity will also attenuate when passing through the medium:

[0041] ;

[0042] Where I represents the intensity of the ultrasonic wave after attenuation, I0 represents the initial intensity of the ultrasonic wave, x represents the distance moved in the medium, and α represents the attenuation coefficient of the medium;

[0043] Step S5: Generate an ultrasonic image through echo.

[0044] Compared with the prior art, the present invention has the following beneficial effects: (1) The present invention takes into account the scattering, absorption and propagation of sound waves in tissues, which helps to simulate the formation of ultrasound images, thereby improving the physical accuracy of the simulation results; (2) The present invention can achieve real-time simulation imaging of tissue and organ movement; (3) The present invention uses statistical technology to simulate soft shadows, and the realism is higher than that of previous real-time ultrasound simulations; (4) The present invention allows simulation of ultrasound imaging in different scenarios, which helps to study the changes in ultrasound images under different conditions, thereby better understanding heart disease. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0046] Figure 1 is a first flow chart of a cardiac dynamic ultrasound simulation method based on statistical ray tracing according to the present invention;

[0047] Figure 2 is a second flow chart of a cardiac dynamic ultrasound simulation method based on statistical ray tracing according to the present invention;

[0048] Figure 3 This is the third flow chart of a cardiac dynamic ultrasound simulation method based on statistical ray tracing according to the present invention. DETAILED DESCRIPTION

[0049] The present invention is further described in detail below to make the purpose, technical solution and advantages of the present invention more clearly understood. Here, the illustrative embodiments of the present invention and their description are used to explain the present invention, but are not intended to limit the present invention.

[0050] like Figure 1 As shown, a cardiac dynamic ultrasound simulation method based on statistical ray tracing includes the following steps:

[0051] Step S1: Initialize the world parameters of the physical simulation computing engine, and define the human tissue material information according to the physical properties of the heart, liver, kidney, blood vessels, lungs, muscles, bones, blood, and skin, wherein the human tissue material information includes acoustic impedance, attenuation coefficient, mirror incidence, roughness, thickness, and expectation and standard deviation of scattering;

[0052] The physical simulation calculation engine uses the Bullet engine, which supports Windows, Linux, MAC, Playstation3, XBOX360, NintendoWii and other platforms. Bullet is also integrated into Maya and Blender 3D, and parameter settings are performed through the Bullet engine, providing a basis for subsequent dynamic cardiac ultrasound simulation.

[0053] Step S2: defining scene file parameters and initializing the scene according to the scene file, wherein the scene file parameters include the position of the ultrasound probe, resolution, pulse stretching rate, sampling angle, sampling number, pitch angle, frequency, transducer spacing, initial intensity of ultrasound, stop threshold, initial medium material of the light, imported model file and model scaling factor;

[0054] Step S3: a method for importing a heart animation model based on Bullet is proposed, wherein the relative position and rotation of each joint in the heart model are calculated through the transformation matrix, and the mesh information of each frame is obtained through the transformation matrix and imported into the Bullet engine, and the mesh is updated according to the time loop;

[0055] like Figure 2 As shown, step S3 also includes:

[0056] Step S31: creating a heart beating animation model;

[0057] Step S32: Use 3D software to analyze the heart model animation and calculate and save the grid of each frame:

[0058] ;

[0059] Among them, Vert ij Indicates the coordinates of the vertex with index j in the i-th frame, globalTrans ij Represents the global transformation matrix of the vertex with index j in the i-th frame, V j Represents the initial coordinates of the vertex with index j;

[0060] Step S33: converting the animation frame grid into grid information that can be used by bullet;

[0061] ;

[0062] Among them, btVert ij Indicates the bullet world vertex position with index j in the i-th frame;

[0063] The 3D software uses FBXSDK, which allows users to access 3D files from most 3D vendors through FBX. The FBX file format supports all major 3D data elements as well as 2D, audio and video media elements. Using FBXSDK to parse the heart model animation can parse multi-modal parameters to ensure the accuracy of the analysis.

[0064] Step S34: importing mesh information into the physical world, and updating the mesh according to the specified time step;

[0065] Step S4: The ultrasonic probe emits rays according to the sampling angle to perform statistical ray tracing; generates random reflection and refraction directions according to the roughness of the surface material of the sound wave hitting point; modifies the hitting point position according to the thickness of the surface material of the hitting point, and generates ultrasonic echo, radio frequency echo, reflection and refraction, and speckle at the new hitting point, and updates the ultrasonic beam intensity at the same time;

[0066] like Figure 3 As shown, step S4 also includes:

[0067] Step S41: Calculate the radio frequency echo composed of the recorded ultrasonic echo:

[0068] ;

[0069] Where t represents time, T represents the time of impact, o represents the source point, and PT represents the impact point, Ω represents the hemispherical surface, I t＞T (t,o) represents the echo contribution received by the source point starting from the intersection point at time T with a direction of ω;

[0070] In practice, it is calculated in a discrete way:

[0071] ;

[0072] Among them, I(r(P T ,ω i )) indicates that by T Towards ω i Directional ray contribution;

[0073] Step S42: Reflection and refraction represent the physical effects that occur when the impact point reaches the boundary of two media. The normal direction is randomly generated according to the roughness to determine the direction of reflection and refraction. At the same time, the reflection and refraction intensity is calculated based on Snell's law:

[0074] ;

[0075] ;

[0076] ;

[0077] ;

[0078] Where θ1 represents the reflection angle, θ2 represents the refraction angle, V i represents the normalized incident direction, N represents the normal direction, Z1 and Z2 represent the acoustic impedance of the two media respectively. V t , V r Indicates the direction of refraction and reflection;

[0079] The intensity update process of the sound wave is as follows:

[0080] ;

[0081] ;

[0082] in, I i Indicates the intensity of the sound wave when it reaches the impact point. I r represents the intensity of the reflected wave, I t It represents the intensity of the refracted wave;

[0083] Step S43: The generation of speckles is due to the fact that countless sub-wavelength particles act as point wave sources to scatter in all directions; the scattering effect is calculated using the ultrasonic speckle convolution model:

[0084] ;

[0085] Among them, x represents the horizontal direction, y represents the axial direction, and z represents the elevation angle. represents the added random noise, r(x,y,z) represents the scattering intensity, g(x,y,z) represents the parameter function of the human tissue scatterer, and h(x,y,z) represents the frequency of the sampler (ultrasound probe). f c Gaussian envelope of the modulation;

[0086] ;

[0087] Among them, σ x , σ y , σ z represent the pulse stretching ratio in transverse, axial and elevation angles respectively;

[0088] Step S44: Update the ultrasonic beam intensity. In addition to the reflection and refraction update, the ultrasonic intensity will also attenuate when passing through the medium:

[0089] ;

[0090] Where I represents the intensity of the ultrasonic wave after attenuation, I0 represents the initial intensity of the ultrasonic wave, x represents the distance moved in the medium, and α represents the attenuation coefficient of the medium;

[0091] S5: Create an image and determine the brightness of the area based on the echo intensity as follows:

[0092] ;

[0093] Among them, I rf Indicates the contribution of the current echo to the image, Intensity indicates the echo intensity, and σ(x,y,z) is expressed as follows:

[0094] When the human tissue scattering probability ξ(x,y,z)≥ρ, where ρ is the scattering probability threshold, the human tissue scattering subvalue, that is, the contribution of the scatterer to the image, is δ(x,y,z)=μ0+ξ(x,y,z)σ, otherwise δ(x,y,z)=0, where μ0 is the scattering expectation and σ is the scattering standard deviation;

[0095] ;

[0096] Among them, row represents the horizontal coordinate of the image, distance represents the distance between the echo and the transducer, c represents the ultrasonic propagation speed, which is defined as 1500m / s, and A r represents the axial resolution;

[0097] ;

[0098] Among them, I [row,i] represents the brightness of the ultrasound image at [row,i], I rf [i] represents the echo contribution belonging to the i-th sampling wave; finally, the image is post-processed, and the image in the polar coordinate system is transformed into the Cartesian coordinate system to obtain the final simulated image.

[0099] Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various equivalent changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A cardiac dynamic ultrasound simulation method based on statistical ray tracing, characterized in that , the method comprises: Step S1: Initialize the world parameters of the physical simulation computing engine, and define the human tissue material information according to the physical properties of the heart, liver, kidney, blood vessels, lungs, muscles, bones, blood, and skin, wherein the human tissue material information includes acoustic impedance, attenuation coefficient, mirror incidence, roughness, thickness, and expectation and standard deviation of scattering; Step S2: defining scene file parameters and initializing the scene according to the scene file, wherein the scene file parameters include the position of the ultrasound probe, resolution, pulse stretching rate, sampling angle, sampling number, pitch angle, frequency, transducer spacing, initial intensity of ultrasound, stop threshold, initial medium material of the light, imported model file and model scaling factor; Step S3: Calculate the relative position and rotation of each joint in the heart model through the transformation matrix, obtain each frame of mesh information through the transformation matrix and import each frame of mesh information into the physical simulation calculation engine, and update the mesh according to the time cycle; Step S4: The ultrasonic probe emits rays according to the sampling angle to perform statistical ray tracing; generates random reflection and refraction directions according to the roughness of the surface material of the sound wave hitting point; modifies the hitting point position according to the thickness of the surface material of the hitting point, and generates ultrasonic echo, radio frequency echo, reflection and refraction, and speckle at the new hitting point, and updates the ultrasonic beam intensity at the same time; Step S5: generating an ultrasonic image through echoes; The step S3 further includes: Step S31: creating a heart beating animation model; Step S32: using 3D software to analyze the heart model animation and calculate and save the mesh of each frame; Step S33: converting the animation frame grid into grid information that can be used by the physical simulation calculation engine; Step S34: importing mesh information into the physical world, and updating the mesh according to the specified time step; The analytical heart model animation is expressed as: Green ij =globalTrans ij *V j Among them, Vert ij Indicates the coordinates of the vertex with index j in the i-th frame, globalTrans ij Represents the global transformation matrix of the vertex with index j in the i-th frame, V j Represents the initial coordinates of the vertex with index j; The expression for converting the animation frame grid into grid information that can be used by the physical simulation calculation engine is: btVert ij =[Translated ij [0],Vert ij [1],Translated ij [2]] Among them, btVert ij Indicates the position of the vertex in the physical simulation computing engine world with index j in the i-th frame, Vert ij [0], Vert ij [1], Vert ij [2] represent the first three coordinates of the quaternion respectively.

2. A cardiac dynamic ultrasound simulation method based on statistical ray tracing as claimed in claim 1, characterized in that: The step S4 further includes: Step S41: Calculate the radio frequency echo composed of the recorded ultrasonic echo; Step S42: Generate a normal direction randomly according to the roughness, determine the reflection and refraction direction, and calculate the reflection and refraction intensity based on Snel I's law: Step S43: generating speckles, calculating the scattering effect through an ultrasonic speckle convolution model; Step S44: Update the ultrasonic beam intensity. In addition to the reflection and refraction updates, the ultrasonic intensity will also attenuate when passing through the medium.

3. A cardiac dynamic ultrasound simulation method based on statistical ray tracing as claimed in claim 2, characterized in that: The calculated radio frequency echo is composed of the recorded ultrasonic echo, and the expression is: I t＞T (t, 0) = ∫ Ω I(t,P T ,ω)cos(ω)dΩ Where t represents time, T represents the time of impact, o represents the source point, and P T represents the impact point, Ω represents the hemispherical surface, I t＞T (t, o) represents the echo contribution received by the source point starting from the intersection point at time T with a direction of ω; Among them, I(r(P T ,ω i )) indicates that by T To ω i The ray contribution in the direction, n represents the number of ultrasonic echoes.

4. A cardiac dynamic ultrasound simulation method based on statistical ray tracing as claimed in claim 2, characterized in that: The reflection and refraction intensity is calculated based on Snell's law, and the expression is: cosθ1=N(-V i ) V r =V i +(2cosθ r )N Where θ1 represents the reflection angle, θ2 represents the refraction angle, and V i represents the normalized incident direction, N represents the normal direction, Z1 and Z2 represent the acoustic impedance of the two media respectively, and V t , V r Indicates the direction of refraction and reflection; The intensity update process of the sound wave is expressed as: Among them, I i Indicates the intensity of the sound wave when it reaches the impact point, I r Indicates the intensity of the reflected wave, I t Represents the intensity of the refracted wave.

5. A cardiac dynamic ultrasound simulation method based on statistical ray tracing as claimed in claim 2, characterized in that: The scattering effect is calculated by the ultrasonic speckle convolution model, and the expression is: r(x,y,z)=g(x,y,z)*h(x,y,z)+∈ Where x represents the lateral direction, y represents the axial direction, z represents the elevation angle, ∈ represents the added random noise, r(x,y,z) represents the scattering intensity, g(x,y,z) represents the parameter function of the human tissue scatterer, and h(x,y,z) is the frequency of the sampler f. c Gaussian envelope of the modulation; Among them, σ x , σ y , σ z Represent the pulse stretching rates in the lateral, axial and elevation angles respectively.

6. A cardiac dynamic ultrasound simulation method based on statistical ray tracing as claimed in claim 2, characterized in that: The updated ultrasonic beam intensity is expressed as: I=I0e -xα Among them, I represents the intensity of the ultrasonic wave after attenuation, I0 represents the initial intensity of the ultrasonic wave, x represents the distance moved in the medium, α represents the attenuation coefficient of the medium, and e represents the natural constant.

Images